Re: Rationale for *text* -> \alert{text} for Beamer export?

[Marcin Borkowski <mbork@wmi.amu.edu.pl>]

Dnia 2013-05-01, o godz. 11:41:49
tsd@tsdye.com (Thomas S. Dye) napisał(a):

> Hi John,
> 
> Jumping in late here, with apologies if that's left me wide of the
> mark.
> 
> John Hendy <jw.hendy@gmail.com> writes:
> 
> > On Wed, May 1, 2013 at 10:00 AM, Marcin Borkowski
> > <mbork@wmi.amu.edu.pl> wrote:
> 
> > Can you explain semantic vs. visual? As in you can more easily
> > customize the meaning of \alert{} or \emph{} whereas \textbf{} and
> > \textit{} only has one meaning? Sort of like using a css tag which
> > can be later customized vs. specifically calling out exactly what
> > you're thinking you want to do at the moment?
> 
> IMHO, the best discussion of this difference is the first chapter of
> Lamport's LaTeX User's Guide and Manual.  Here is the gist as I
> understand it:
> 
> 1) A principle of typesetting is that the layout of a document should
> reflect its logical structure.
> 
> 2) A computer typesetting program can achieve this if it knows what
> key parts of the document mean.
> 
> 3) So, markup should be semantic, rather than visual.
> 
> It is possible to achieve identical results using visual markup, of
> course, but why not let the computer keep track of things instead?

+1

Notice also that even LaTeX breaks the rule of "use only semantic
markup in the document" (and in fact, there are cases when the rule is
a bit fuzzy anyway).  Finding examples of /visual/ markup in LaTeX
(without semantic counterparts) are left as an exercise for the
reader;).

(Hint: rahzrengvbaf fglyrf qrcraq abg ba gur punenpgre bs gur
rahzrengvba, ohg ba vgf qrcgu, naq jvgubhg cnpxntrf yvxr rahzvgrz vg'f
abg rnfl gb qrsvar lbhe bja rahzrengvba fglyrf.)

> > Sure, and understood. In general, I'm using *text* simply to call
> > attention to something important. I work in product development, so
> > something like:
> >
> > Customer response to product sampling:
> > - *US:* blah blah blah
> > - *China:* blah blah blah
> > - *India: blah blah blah
> 
> Here, to achieve semantic markup, you would use description lists
> 
> - US :: blah
> - China :: blah
> - India :: blah
> 
> The :: separator lets Org (and ultimately LaTeX) know that the part
> before the separator is the term that is being described.

And then use the enumitem package to customize the exact look of the
description environment.

> hth,
> Tom

Regards,

-- 
Marcin Borkowski
http://octd.wmi.amu.edu.pl/en/Marcin_Borkowski
Adam Mickiewicz University